Method and system for automatically setting running program of household appliance and household appliance

Abstract

A method and system for automatically setting a running program of a household electrical appliance, and the household electrical appliance, the method comprising: in response to detecting that a preset trigger condition is satisfied, obtaining a current time, the preset trigger condition being used for triggering an automatic setting running program; the current time is input into a preset machine learning model, a current running program is obtained, the preset machine learning model is used for predicting the current running program according to the current time and historical running programs of the household appliance at related time in history, and the related time is associated with the current time; and setting the household appliance to run according to the current running program. Through the scheme of the invention, the running program meeting the user demand at the current time can be more accurately and automatically set when the household electrical appliance runs this time, and the intelligent degree of the household electrical appliance and the use satisfaction degree of the user are further improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of household appliances, and in particular, to a method, a system and a household appliance for automatically setting an operating program of a household appliance. Background Art

[0002] With the development of household appliance technology, various household appliances are increasingly tending towards intelligent and personalized designs to meet the increasingly high requirements of users for operation convenience. A major feature of existing household appliances is that they can automatically set the operating program when powered on, without the need for users to set it manually, in order to improve the user's usage convenience and provide a better experience.

[0003] Taking a washing machine as an example, at present, some washing machines have a memory mode and default to running according to the operating program set at the previous power-on each time they are powered on. However, in practice, the simple memory mode cannot well meet the user's needs. For example, when the user set the operating program to quick wash to shorten the washing time at the previous power-on, and needs to wash delicate clothes such as woolen sweaters at this power-on, it is obviously inappropriate for the washing machine to automatically set the operating program to quick wash according to the memory mode, and the user still needs to manually change the operating program of the washing machine to gentle wash. This operation method that still requires user intervention obviously goes against the original intelligent design intention of the washing machine.

[0004] Therefore, there is an urgent need to provide a more intelligent automatic operating program setting scheme that can truly "think what the user thinks" and more accurately automatically set the operating program that meets the user's current usage needs for the user each time the household appliance runs. Summary of the Invention

[0005] An object of the embodiments of the present invention is to provide an improved method, a system and a household appliance for automatically setting an operating program of a household appliance.

[0006] Therefore, the embodiments of the present invention provide a method for automatically setting an operating program of a household appliance, including: in response to detecting that a preset trigger condition is satisfied, obtaining the current time, where the preset trigger condition is used to trigger the automatic setting of the operating program; inputting the current time into a preset machine learning model to obtain the current operating program, where the preset machine learning model is used to predict the current operating program according to the current time and the historical operating programs of the household appliance at related times in the past, and the related times are associated with the current time; setting the household appliance to run according to the current operating program.

[0007] Compared with the existing household appliances that automatically set the operating program using the memory mode, which is not intelligent enough and the accuracy of the automatically set operating program matching the actual needs of users is low, this implementation can more accurately automatically set the operating program that meets the needs of users at the current time during the current operation of the household appliance, further improving the intelligence level of the household appliance and the satisfaction of users. Specifically, the current time when the household appliance meets the preset trigger condition is used as one of the consideration factors and input into the preset machine learning model, and the current operating program is predicted through the processing and operation of the preset machine learning model. The preset trigger condition can be, for example, the household appliance is powered on, or it can be, for example, the function of automatically setting the operating program of the household appliance is enabled. Further, based on the preset machine learning model, fully considering the different styles / habits of users when using household appliances at different times, predicting the usage needs of users at the current time according to the operation history of users at relevant times, and then automatically setting the corresponding current operating program. Thus, it is possible to set the operating program more intelligently after power-on to more accurately provide the user with the operating program that meets the needs of the user at the current time.

[0008] Further, when the preset machine learning model makes a prediction, it comprehensively considers the historical operating programs set by the household appliance at relevant times in the past. Thus, the preset machine learning model can learn and remember the user's behavior and usage habits according to the user's historical usage records, predict the usage needs of the user when operating the household appliance this time based on historical data, and thus provide a more appropriate personalized operating program for the user.

[0009] Further, the preset machine learning model predicts and outputs the current operating program through historical data. As the historical data continues to expand, the household appliance can become more and more intelligent as the user uses it. When the user's usage habits change, the household appliance can also make timely adjustments. For household appliances of different users, as the usage habits of different users are different, the differences in the prediction results of each household appliance at the same current time will become larger and larger. Thus, adopting this implementation can allow users to personalize the setting of the operating program of the household appliance, and this kind of personalized setting is achieved almost without the user's perception and without the need for the user to manually set it, greatly improving the intelligence level of the household appliance and being beneficial to optimizing the user experience.

[0010] Optionally, the correlation between the relevant time and the current time includes: the relevant time and the current time have a preset time interval, and the preset time interval includes a calendar week. Thus, by recording the operating programs set by the user when using household appliances every day in a cycle of weeks, the differences in the user's usage habits of household appliances at different times can be reflected, and the current operating program can be set more pertinently to more intelligently adapt to the user's usage needs at the current time. Further, when the user's usage habits have a periodic change trend, the periodic characteristics are reflected through the preset time interval, so that the predicted current operating program conforms to the periodic change trend. Further, the preset time interval including the calendar week can also adapt to the work, life rhythms and habits of most users.

[0011] Optionally, different times correspond to different preset machine learning models. The step of inputting the current time into the preset machine learning model to obtain the current operating program includes: determining the corresponding preset machine learning model according to the current time; inputting the current time into the determined preset machine learning model to obtain the current operating program. Thus, using their respective preset machine learning models for training and prediction at different times is beneficial to improving the accuracy of model prediction. Further, each model is trained and iteratively updated with the historical data of its respective relevant time, so that each model is more focused on predicting the corresponding operating program according to the user's usage habits at its respective corresponding time.

[0012] Optionally, the number of the preset machine learning models is 7, corresponding to each day of the calendar week respectively. Thus, each of the seven days of a week corresponds to one of the 7 preset machine learning models, and each preset machine learning model can be focused on predicting the operating program required by the user when using household appliances on that day.

[0013] Optionally, the preset machine learning models corresponding to different times are the same when the household appliance is powered on for the first time in history, and independently change as the household appliance is used. In other words, initially, the same preset machine learning model is defaulted every day, and as the user uses it, differences gradually arise according to the user's usage habits. Thus, the process of program burning in the factory setting of the household appliance is simplified, which is beneficial to reducing the manufacturing cost and process complexity. Further, as the user uses it, each model gradually accumulates historical data for learning and iterative update according to the program setting habits of the user when using it at the corresponding time, so as to realize the intelligent and personalized setting of the operating program according to the user's usage.

[0014] Optionally, the preset machine learning model is iteratively updated as the number of times the household appliance is used increases. Thus, continuously training the preset machine learning model with the accumulation of user usage data, the prediction of the preset machine learning model will be more targeted and more able to reflect the user's personalization.

[0015] Optionally, the iterative update process of the preset machine learning model includes: receiving feedback data, where the feedback data is used to characterize the set satisfaction degree for the historical running program; constructing a training set and a validation set based on the feedback data received within a period of time and the corresponding historical running program; training the preset machine learning model based on the training set to obtain an updated preset machine learning model; and validating the updated preset machine learning model based on the validation set. Thus, according to the user's satisfaction feedback on the program settings when using household appliances in the past, the model is automatically optimized and improved to automatically set a more appropriate and accurate running program after the next startup, significantly improving user satisfaction.

[0016] Optionally, the feedback data includes satisfaction scores for at least one parameter selected from: dryness, noise, entanglement degree, and the correctness of the automatic setting of the running program. Thus, the preset machine learning model can more specifically optimize and update the areas where the user is dissatisfied or not very satisfied.

[0017] Optionally, the method for automatically setting the running program of a household appliance further includes: sending a prompt message, where the prompt message includes a feedback form; receiving the feedback form and generating the feedback data based on the feedback form. Thus, by actively prompting the user to submit feedback, as much feedback data as possible is collected for training the model to optimize the model training effect.

[0018] Optionally, the prompt message and / or the feedback form are transmitted through the display and / or input unit of the household appliance. Thus, the user is reminded to give feedback through the human-computer interaction interface provided on the household appliance, and the feedback data is received as the basis for model training.

[0019] Optionally, the prompt message and / or the feedback form are transmitted through a terminal device associated with the household appliance. Thus, the user can receive the feedback form and upload the feedback data anytime and anywhere, and the feedback process is more convenient.

[0020] Therefore, an embodiment of the present invention further provides a household appliance, including: a body; a control module disposed in the body, where the control module is used to execute the method for automatically setting the running program of the foregoing household appliance. Thus, the iterative update actions of the preset machine learning model are all completed within the household appliance, so that the household appliance itself can complete model optimization with a fast response speed. Further, the iterative update actions of the preset machine learning model do not require the support of an external server and do not depend on the network, and the model optimization can be achieved even if the household appliance is not connected to the network.

[0021] Optionally, the household appliance is selected from: a washing machine, a washer-dryer, and a dryer. Thus, in view of the significant differences in the styles of washing and caring for clothes by users at different times (e.g., weekdays and weekends), the method described in this embodiment is applied to household appliances such as washing machines, enabling the washing machines to provide more intelligent washing and caring functions for users. Further, in addition to washing machines, the method for automatically setting the operating program of the household appliances described in this embodiment can also be extended and applied to other washing / drying household appliances, making the entire process from washing to drying clothes more intelligent and convenient.

[0022] Optionally, the household appliance further includes: a communication module disposed on the body, and the control module receives feedback data through the communication module; and / or, a display and / or input unit disposed on the body, and the control module receives feedback data through the display and / or input unit. Thus, users can either transmit satisfaction through network communication or directly input feedback information through the display and / or input unit, improving the convenience of feedback.

[0023] Therefore, an embodiment of the present invention further provides an automatic operating program setting system for a household appliance, including: a household appliance, including a body and a control module, where the control module is configured to execute the method for automatically setting the operating program of the household appliance described above; a server communicatively connected to the control module, and the server is configured to synchronize the preset machine learning model to the control module. Thus, the server can be used to store historical data, including historical operating programs and corresponding feedback data, and retrain the model based on the historical data, so that the currently predicted operating program of the household appliance based on the updated model when it is powered on next time is more in line with the user's usage habits at the current time. Further, when the server is externally provided to the household appliance, it is beneficial to reduce the number of components inside the household appliance and lower the cost.

[0024] Optionally, the server is configured to iteratively update the preset machine learning model based on the received historical operating program and feedback data of the household appliance, and synchronize the updated preset machine learning model to the control module. Thus, with the support of the high computing power of the server, the iterative response speed of the preset machine learning model can be further improved.

[0025] Optionally, the automatic operating program setting system for the household appliance further includes: a communication module disposed on the body, and the control module communicates with the server through the communication module. Thus, the household appliance establishes a communication connection with the outside world (e.g., the server) through the communication module to achieve remote update of the preset machine learning model.

[0026] Optionally, the operation program automatic setting system of the household appliance further includes: a display and / or input unit, which is arranged on the body and communicates with the control module, and the display and / or input unit is used to receive feedback data. Thus, the user can submit feedback data locally on the household appliance, and the household appliance summarizes it together with the corresponding historical operation program to the server side, so that the server can iteratively update the preset machine learning model based on these data. Description of the Drawings

[0027] Figure 1 is a flowchart of a method for automatically setting an operation program of a household appliance according to an embodiment of the present invention;

[0028] Figure 2 is Figure 1 a flowchart of a specific implementation manner of step S12 in

[0029] Figure 3 is a flowchart of an iterative update process of a preset machine learning model according to an embodiment of the present invention;

[0030] Figure 4 is a schematic diagram of a household appliance according to an embodiment of the present invention;

[0031] Figure 5 is a schematic diagram of the principle of an operation program automatic setting system of a household appliance according to an embodiment of the present invention;

[0032] Figure 6 is a schematic diagram of the process of automatically setting an operation program of a household appliance in a typical application scenario according to an embodiment of the present invention;

[0033] In the drawings:

[0034] 1 - Household appliance; 11 - Body; 12 - Control module; 13 - Communication module; 14 - Display and / or input unit; 15 - Power-on button; 16 - Door; 2 - Server; 3 - Terminal device. Detailed Embodiment

[0035] As mentioned in the background art, in the prior art, the program automatic setting method of household appliances adopts a memory mode, and each time it is powered on, it defaults to run according to the operation program set at the previous power-on. Since it cannot truly "think what the user thinks", the existing household appliances are not intelligent enough.

[0036] The inventors of the present application analyzed and found that one of the reasons for the above problems is that the existing program automatic setting scheme based on the memory mode ignores the differences in user usage requirements each time the household appliance is used.

[0037] Taking a washing machine as an example, existing washing machines with a memory mode ignore the differences in the number, material, color, thickness, etc. of the laundry items each time, and are not intelligent enough. Specifically, according to the different activities carried out every day within a week, the laundry needs of users are also different every day. For example, on Monday, the user needs a sports wear (the characteristics can be, for example, lightweight clothing) program, and on Tuesday, the user needs a work wear (the characteristics can be, for example, easily deformed clothing) program, etc. In this case, the simple memory mode obviously cannot meet the personalized needs of users every day, resulting in the user still having to manually debug the correct operating program every time it is used.

[0038] In addition, in the prior art, there is no feedback channel for users' satisfaction with the automatic program setting of household appliances, and household appliances cannot adaptively optimize, update, and iterate their own program automatic setting logic according to user feedback.

[0039] To solve the above technical problems, an embodiment of the present invention provides a method for automatically setting an operating program of a household appliance, including: in response to detecting that a preset trigger condition is satisfied, obtaining the current time, where the preset trigger condition is used to trigger the automatic setting of the operating program; inputting the current time into a preset machine learning model to obtain a current operating program, where the preset machine learning model is used to predict the current operating program according to the current time and the historical operating programs of the household appliance at related times in the past, and the related time is associated with the current time; setting the household appliance to operate according to the current operating program.

[0040] Thus, this implementation scheme can more accurately automatically set an operating program that meets the user's needs at the current time during the current operation of the household appliance, further improving the intelligence level of the household appliance and the user's satisfaction. Specifically, taking the current time when the household appliance satisfies the preset trigger condition as one of the consideration factors and inputting it into the preset machine learning model, the current operating program is predicted through the processing and operation of the preset machine learning model. The preset trigger condition can be, for example, the household appliance is powered on, or it can also be, for example, the automatic setting operating program function of the household appliance is enabled. Further, based on the preset machine learning model, fully considering the different styles / habits of users when using household appliances at different times, predicting the user's usage needs at the current time according to the user's operation history at related times, and then automatically setting the corresponding current operating program. Thus, it is possible to more intelligently set the operating program after power-on to more accurately provide the user with an operating program that meets the user's needs at the current time.

[0041] Further, when the preset machine learning model makes a prediction, it comprehensively considers the historical operating programs set by the household appliance at relevant times in the past. Thus, the preset machine learning model can learn and remember the user's behavior and usage habits based on the user's historical usage records, predict the user's usage needs when operating the household appliance this time according to the historical data, and thus provide a more appropriate personalized operating program for the user.

[0042] Further, the preset machine learning model predicts and outputs the current operating program through historical data. As the historical data continues to expand, the household appliance can become more and more intelligent as the user uses it. When the user's usage habits change, the household appliance can also make adjustments in a timely manner. For household appliances of different users, as the usage habits of different users are different, the differences in the prediction results of each household appliance at the same current time will become larger and larger. Thus, adopting this implementation scheme can allow users to customize the operating program of the household appliance, and this kind of customization can be achieved without the user's manual setting almost imperceptibly to the user, greatly improving the intelligence level of the household appliance and being beneficial to optimizing the user experience.

[0043] To make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0044] Figure 1 It is a flowchart of a method for automatically setting an operating program of a household appliance according to an embodiment of the present invention.

[0045] A household appliance can be pre-set with at least one operating program (program). When operating according to different operating programs, the operating parameters of at least one component in the household appliance are different. The operating program adopted by the household appliance for this operation can be set by the user or adaptively set by the household appliance according to this implementation scheme.

[0046] This implementation scheme can be applied to the scenario of automatically setting the operating program of a household appliance, and the household appliance can be, for example, a washing machine. The operating program can include parameters such as the water intake, water inlet temperature, drum speed of the washing machine, as well as the execution duration and the order of execution of each operation step (such as the rinsing, dehydration steps, etc.).

[0047] This implementation scheme can be executed by a control module. The control module can be, for example, a single-chip microcomputer of the household appliance, or can also be, for example, a control unit dedicated to executing this implementation scheme in the household appliance. In this embodiment, the control module can be arranged in the body of the household appliance.

[0048] In some embodiments, the household appliance can be, for example, a washing machine, a washer-dryer, a dryer, an air conditioner, and a cooker, etc. Next, this implementation scheme will be elaborated in detail taking a washing machine as an example.

[0049] Specifically, referring to Figure 1 , the method for automatically setting the operating program of the household appliance according to this embodiment may include the following steps:

[0050] Step S11, in response to detecting that a preset trigger condition is satisfied, obtain the current time, where the preset trigger condition is used to trigger the automatic setting of the operating program;

[0051] Step S12, input the current time into a preset machine learning model to obtain the current operating program, where the preset machine learning model is used to predict the current operating program according to the current time and the historical operating programs of the household appliance at relevant times in the past, and the relevant times are associated with the current time;

[0052] Step S13, set the household appliance to operate according to the current operating program.

[0053] In some embodiments, the preset trigger condition may be, for example, the startup of the household appliance. The startup of the household appliance may include the power-on of the household appliance, and further may include the switching of the household appliance from the sleep state / standby state to the working state.

[0054] Furthermore, the startup action of the household appliance may be triggered in response to a user operation. For example, the washing machine may be provided with a power-on button 15 on the front panel (as Figure 4 shown). When the user presses or touches the power-on button 15, the washing machine starts up. Correspondingly, when the control module detects this startup action, it executes step S11 and records the current time as the startup time of this washing machine.

[0055] For another example, the startup action may be automatically triggered in response to detecting that the user opens the door 16 of the washing machine (as Figure 4 shown), so as to further improve the intelligence level and use convenience of the washing machine.

[0056] Alternatively, the startup action of the household appliance may be triggered regularly. For example, the user pre-sets a startup time, and when the startup time arrives, the household appliance automatically starts up.

[0057] Furthermore, after the washing machine completes startup, for example, after the user presses or touches the power-on button, a startup signal may be generated. The control module 12 (as Figure 4 shown) obtains the current time after receiving the startup signal. The current time may be the time data generated after receiving the startup signal, or may be the time data generated when the washing machine starts to execute the operating program (for example, when the washing machine starts to fill with water).

[0058] In some embodiments, the preset trigger condition can also be, for example, that the automatic setting operation program function of the household appliance is enabled. Specifically, the automatic setting operation program can be a function provided by the household appliance, and the user can choose whether to use this function as needed. For example, a quick button can be set on the washing machine. In response to the triggering of the quick button, the control module determines that it has detected the user's use of the automatic setting operation program function, that is, it has detected that the preset trigger condition is met.

[0059] In some embodiments, the current time can be obtained from a clock module built into the washing machine. Alternatively, the clock module can also be externally disposed on the household appliance and communicate with the control module.

[0060] Furthermore, the preset machine learning model can receive the input current time and predict the current operation program based on the received current time and the historical operation programs executed by the household appliance at relevant times in the past. Correspondingly, the control module sets the specific operation parameters of the household appliance according to the prediction result of the preset machine learning model.

[0061] Furthermore, the user's usage habits of using the household appliance at the relevant time and the current time are basically the same. Correspondingly, the historical operation program set when powering on at the relevant time has a high probability of being adapted to the current operation program that needs to be set when the household appliance is powered on this time. Further, the historical operation program can include the operation program automatically set when powering on at the relevant time in the past. Specifically, the operation program automatically set at the relevant time in the past can be, for example, the operation executed by the washing machine at the relevant time in the past Figure 1 The solution of the illustrated embodiment, the prediction result output by the preset machine learning model. Alternatively, the historical operation program can also be the operation program set by the user at the relevant time in the past.

[0062] As described above, by adopting this implementation solution, it is possible to more accurately automatically set the operation program that meets the user's needs at the current time during the current operation of the household appliance, further improving the intelligence level of the household appliance and the user's satisfaction. Specifically, the current time when the household appliance meets the preset trigger condition is used as one of the consideration factors and input into the preset machine learning model. After being processed and calculated by the preset machine learning model, the current operation program is predicted. The preset trigger condition can be, for example, the power-on of the household appliance, or it can also be, for example, that the automatic setting operation program function of the household appliance is enabled. Further, based on the preset machine learning model, fully considering the different styles / habits of users when using household appliances at different times, predicting the user's usage needs at the current time according to the user's operation history at the relevant time, and then automatically setting the corresponding current operation program. Thus, it is possible to more intelligently set the operation program after power-on to more accurately provide the operation program that meets the user's needs at the current time for the user.

[0063] Furthermore, when the preset machine learning model makes a prediction, it comprehensively considers the historical operation programs set by the household appliance at relevant times in the past. Thus, the preset machine learning model can learn and remember the user's behavior and usage habits based on the user's historical usage records, predict the usage requirements of the user when operating the household appliance this time according to the historical data, and thus provide a more appropriate personalized operation program for the user.

[0064] Furthermore, the preset machine learning model predicts and outputs the current operation program through historical data. As the historical data continues to expand, the household appliance can become more and more intelligent as the user uses it. When the user's usage habits change, the household appliance can also make timely adjustments. For household appliances of different users, as the usage habits of different users are different, the differences in the prediction results of each household appliance at the same current time will become larger and larger. Thus, adopting this implementation solution can allow users to set the operation programs of household appliances personalizedly, and this kind of personalized setting can be achieved without the user's manual setting almost imperceptibly by the user, greatly improving the intelligence level of the household appliance and being beneficial to optimizing the user experience.

[0065] In a specific implementation, the correlation relationship between the relevant time in the past and the current time may include: there is a preset time interval between the relevant time and the current time, and the preset time interval includes the calendar week.

[0066] Specifically, in this implementation solution, the measurement unit of time can be days. For example, assuming the current time is Monday, correspondingly, the relevant time can be Monday of each week in the past. The preset machine learning model can predict the current operation program that should be automatically set this Monday according to the historical operation programs set when the household appliance was turned on every Monday in the past.

[0067] Thus, recording the operation programs set by the user every day when using the household appliance in a weekly cycle can reflect the differences in the user's usage habits of the household appliance at different times, and set the current operation program more pertinently to more intelligently adapt to the user's usage requirements at the current time. Further, when the user's usage habits have a periodic change trend, the periodic characteristics are reflected through the preset time interval, so that the predicted current operation program conforms to the periodic change trend. Further, the preset time interval includes the calendar week, which can also adapt to the work, life rhythms and habits of most users.

[0068] In some embodiments, it is possible to record the operation programs set each time the preset trigger condition (for example, each time it is turned on) is met in the past and the corresponding trigger times (for example, the turn-on time).

[0069] In step S12, when the preset machine learning model makes a prediction, it can determine the associated relevant time based on the current time, and then determine the historically recorded running program with the startup time as the relevant time as the historical running program for reference of this model prediction.

[0070] In a variation, the specific interval size of the preset time interval can be adjusted according to the user's living habits and usage habits of household appliances. For example, if the user's usage habits of household appliances change every three calendar days, the preset time interval can be set to three calendar days.

[0071] In practical applications, the preset time interval can be uniformly set to one calendar week (ie, 7 calendar days) when the household appliance leaves the factory. The user can adjust the specific value of the preset time interval as needed during the use of the household appliance.

[0072] In one specific implementation, different times may correspond to different preset machine learning models.

[0073] Specifically, each day within a single preset time interval may have a corresponding preset machine learning model.

[0074] Taking the preset time interval as a calendar week as an example, the number of the preset machine learning models can be 7, corresponding to each day of the calendar week. Thus, the seven days of the week correspond to one of the 7 preset machine learning models, and each of the preset machine learning models can focus on predicting the running program required by the user when using the household appliance on that day.

[0075] In one specific implementation, the preset machine learning models corresponding to different times can be the same when the household appliance is turned on for the first time in history, and change independently as the household appliance is used.

[0076] Specifically, the first time the household appliance is powered on in history may include the first time the household appliance is powered on after leaving the factory, and may also include the first time the household appliance is powered on by a user in history.

[0077] Furthermore, home appliances can be pre-set with 7 identical preset machine learning models when they leave the factory, and each preset machine learning model is iteratively updated according to the user's usage habits at the corresponding time. In other words, the same preset machine learning model is used by default every day at the beginning, and as users use it, differences gradually occur according to user usage habits.

[0078] Thus, the process of program burning in the factory settings of household appliances is simplified, which helps reduce the manufacturing cost and process complexity. Further, as the user uses the appliance, each model gradually accumulates historical data according to the user's program setting habits during use at corresponding times for learning and iterative update, so as to realize the intelligent and personalized setting of the operating program according to the user's use. In some embodiments, the control module may have a memory function and can actively collect the current time of each startup operation of the household appliance and the set current operating program as the training basis for the preset machine learning model, so that the preset machine learning model can more accurately predict the current operating program to be executed at the next current time.

[0079] Specifically, the collected data may include multiple data groups, where each data group includes corresponding current time data and current operating program data. These data groups can form a training set. After inputting the training set into the preset machine learning model, the preset machine learning model learns the corresponding relationships in the data groups for predicting the current operating program to be executed at the next current time.

[0080] It should be noted that the relevant time corresponding to each current time is the "current time" input into the preset machine learning model in history. Correspondingly, the current time input into the preset machine learning model by the method shown this time becomes the relevant time of the next current time when the household appliance is turned on again after a preset time interval. Figure 1 The current time input into the preset machine learning model by the method shown this time becomes the relevant time of the next current time when the household appliance is turned on again after a preset time interval.

[0081] Further, multiple data groups can be further divided at preset time intervals to respectively train / iteratively update the corresponding preset machine learning models. For example, all data groups composed of each Monday in history and the corresponding historical operating programs can be input into the preset machine learning model corresponding to Monday for model training. Another example is that all data groups composed of each Wednesday in history and the corresponding historical operating programs can be input into the preset machine learning model corresponding to Wednesday for model training.

[0082] Since the data groups input into different preset machine learning models are different, the corresponding relationships analyzed by their operations are also different. Then different corresponding relationships can be used according to different current times to output different current operating programs.

[0083] Further, as the number of data groups accumulated in each preset machine learning model gradually increases, the differences between the preset machine learning models will also gradually increase, and this difference is consistent with the differences in the user's usage requirements for household appliances at different times.

[0084] In a specific implementation, referring to Figure 2 , step S12 may include the following steps:

[0085] Step S121, determine the corresponding preset machine learning model according to the current time;

[0086] Step S122, input the current time into the determined preset machine learning model to obtain the current running program.

[0087] For example, a washing machine can be preset with 7 preset machine learning models, corresponding to Monday to Sunday respectively. When the user starts the washing machine on a certain Monday, that Monday will be used as the current time. Call the preset machine learning model corresponding to Monday according to the current time of Monday. Further, input Monday into the preset machine learning model called this time. The preset machine learning model predicts and outputs the current running program to be executed this Monday according to the historical running program set when starting up on Monday in the past.

[0088] Thus, using their respective preset machine learning models for training and prediction at different times is beneficial to improving the accuracy of model prediction. Further, each model is trained and iteratively updated with the historical data of its respective relevant time, so that each model is more focused on predicting the corresponding running program according to the user usage habits of its respective corresponding time.

[0089] In a specific implementation, the preset machine learning model can be iteratively updated as the number of times of using the household appliance increases. Specifically, the current time and the set current running program each time the household appliance is turned on can be added to the historical data for retraining the preset machine learning model. Thus, continuously training the preset machine learning model with the accumulation of user usage data, the prediction of the preset machine learning model will be more targeted and more able to reflect user personalization.

[0090] Taking the washing machine as an example, the washing machine can record the current time and the current running program of each startup. The recorded current time corresponds one-to-one with the recorded current running program. The data groups containing the current time and the current running program further form a historical data set. The number of elements contained in the historical data set increases with each use of the washing machine, and the historical data set is thus continuously updated and expanded.

[0091] As the historical data set is updated and expanded, the preset machine learning model can be continuously iteratively updated accordingly.

[0092] In some embodiments, for different preset machine learning models associated with different times, different subsets of the above historical data set can be used for training respectively. Among them, different subsets can include data groups corresponding to different times within a preset time interval.

[0093] In some embodiments, machine learning algorithms can be used to construct a preset machine learning model. The machine learning algorithms can be, for example, decision trees, naive Bayes classification, least squares regression, logistic regression, support vector machines, neural networks, deep learning, etc.

[0094] In a specific implementation, as Figure 3 shown, the iterative update process of the preset machine learning model can include the following steps:

[0095] Step S123, receiving feedback data, where the feedback data is used to characterize the set satisfaction with respect to the historical running program;

[0096] Step S124, constructing a training set and a validation set based on the feedback data received within a period of time and the corresponding historical running program;

[0097] Step S125, training the preset machine learning model based on the training set to obtain an updated preset machine learning model;

[0098] Step S126, validating the updated preset machine learning model based on the validation set.

[0099] Specifically, in step S123, the feedback data can come from the user. For example, the feedback data input by the user can be received through a display and / or input unit.

[0100] Furthermore, the feedback data can include satisfaction evaluations of at least one parameter, and the parameter can be selected from: dryness, noise, twine degree, and the correctness of the automatic setting of the running program. For example, the satisfaction with any of the above parameters can be characterized in the form of scoring. The twine degree can refer to the twine degree of the clothes.

[0101] Taking a drum washing and drying integrated washing machine as an example, the factors that may affect the user experience and the laundry effect include the dryness of the washed clothes, the noise generated during the laundry process, the twine degree of the clothes, and the correctness of the automatic setting of the running program, etc.

[0102] Thus, the preset machine learning model can more specifically optimize and update the areas where the user is dissatisfied or not very satisfied.

[0103] Further, by statistically analyzing the feedback data, it is possible to determine whether the setting of at least one parameter in the historical operation program reflected by the feedback data is reasonable, and then adjust the specific value of the parameter with an unreasonable setting. For example, historically, the historical operation program set after meeting the preset trigger condition on Tuesday is gentle wash. After the household appliance finishes running, the user uploads feedback data, and the feedback data reflects dissatisfaction with the noise during the operation of the household appliance. Correspondingly, in step S125, when updating the preset machine learning model, this feedback data can be considered, and the rotation speed of the drum during gentle wash on Tuesday can be reduced in order to reduce the noise.

[0104] Thus, when predicting the current operation program at the current time, the preset machine learning model determines a current operation program that better meets the user's requirements, or at least the specific values of at least one parameter in the current operation program are more accurate and to the user's liking.

[0105] In some embodiments, after the current operation program is executed, the household appliance can also perform the steps of: sending a prompt message, where the prompt message includes a feedback form; receiving the feedback form, and generating the feedback data based on the feedback form.

[0106] For example, after each laundry is completed, the washing machine can send a prompt message to the user through an interactive panel, such as displaying a feedback form on the UIM. The feedback form includes options for at least one parameter for the user to evaluate the above parameters. For example, the options can be satisfied / suitable (e.g., can be recorded as 1) or dissatisfied / inappropriate (e.g., can be recorded as 0).

[0107] In a variant, the satisfaction level can also be in the form of a score. For example, from dissatisfied to satisfied, it corresponds to 1 - 5 points, dissatisfied is 1 point, and satisfied is 5 points.

[0108] After the user fills out the feedback form, they can click to upload. In response to receiving the feedback form, the control module can generate feedback data according to the evaluations of each parameter in the feedback form.

[0109] Thus, by actively prompting the user to submit feedback, as much feedback data as possible is collected for training the model, optimizing the model training effect.

[0110] In a specific embodiment, the prompt information and / or the opinion feedback form can be transmitted through the display and / or input unit of the household appliance. The display and / or input unit can be a User Interface Module (UIM for short). For example, after this laundry is finished, a prompt information is popped up on the UIM to prompt the user to give feedback. In response to the user clicking to agree to give feedback, the control module further displays an opinion feedback form on the UIM for the user to fill out. Thus, the user is reminded to give feedback through the human-machine interaction interface provided on the household appliance, and feedback data is received to be used as the basis for model training.

[0111] In a specific implementation, the prompt information and / or the opinion feedback form can be transmitted through a terminal device associated with the household appliance. For example, after this laundry is finished, the control module can send a message indicating that the running program has been completed to the terminal device, and the terminal device sends the prompt information and receives the opinion feedback form submitted by the user. Thus, the user can receive the opinion feedback form and upload feedback data anytime and anywhere, and the feedback process is more convenient.

[0112] In a specific implementation, the prompt information and the opinion feedback form can be displayed or received through different media. For example, the prompt information can be sent through the display and / or input unit of the household appliance, and the opinion feedback form is received on the user's mobile phone.

[0113] In some embodiments, the display and / or input unit can be arranged on the body of the washing machine and can communicate with the terminal device through a wireless communication method. The wireless communication method can be, for example, Wireless Fidelity (WIFI for short) and Near Field Communication (NFC for short), etc.

[0114] Specifically, the terminal device can include a mobile terminal device, such as the user's mobile phone, tablet computer, laptop computer, etc. It can also be, for example, other smart home appliances located in the same local area network as the household appliance, such as the refrigerator, range hood, etc. in the user's home.

[0115] Further, the opinion feedback form is transmitted to the user's mobile phone through a wireless signal. After the user scores and evaluates according to their own satisfaction, the terminal device can further extract feedback data from the opinion feedback form submitted by the user, and then transmit it back to the washing machine or to the cloud server through wireless transmission.

[0116] In a specific implementation, in step S124, the data groups and feedback data in the historical dataset can be corresponded to obtain the historical dataset. Specifically, the historical dataset includes multiple data groups, and each data group includes three data with corresponding relationships to each other: time data, running program data, and feedback data.

[0117] Furthermore, the historical dataset can be divided according to a certain ratio to obtain a training set and a validation set. The training set can be used to generate a preset machine learning model, and then the validation set is used to verify whether the parameters of the preset machine learning model obtained by training are appropriate.

[0118] In some embodiments, since the training set and the validation set play different roles in the process of iteratively updating the preset machine learning model, the requirements for the number of elements in the set can also be different. Usually, the training set contains more data groups than the validation set, and the ratio between the two can be, for example, 6 (training set): 4 (validation set), or can also be, for example, 7:3.

[0119] Furthermore, if the verification result of step S126 indicates that the accuracy of the prediction result of the updated preset machine learning model is higher than the preset threshold, it can be confirmed that the updated preset machine learning model is appropriate. Correspondingly, the updated preset machine learning model can be used for the next execution Figure 1 the scheme shown.

[0120] If the verification result of step S126 indicates that the accuracy of the prediction result of the updated preset machine learning model is lower than the preset threshold, it indicates that the updated preset machine learning model is not appropriate. Correspondingly, steps S125 and S126 can be re-executed to retrain the preset machine learning model until the accuracy of the prediction result of the updated preset machine learning model for the time data in the validation set and the corresponding running program data in the validation set is higher than the preset threshold.

[0121] In some embodiments, when re-executing step S125 and step S126, the training set and the validation set can be re-divided.

[0122] Furthermore, in an example where different preset machine learning models correspond to different times, the historical dataset can be applicable to all preset machine learning models. Each preset machine learning model can be trained and iteratively updated according to the historical running programs and feedback data corresponding to their respective corresponding times.

[0123] Thus, according to the user's satisfaction feedback on the program settings when using household appliances historically, the model is automatically optimized and improved to automatically set a more appropriate and accurate running program after the next startup, significantly improving user satisfaction.

[0124] In a variant, different preset machine learning models may correspond to different historical data sets. Correspondingly, each data group in the historical data set may include a historical operation program and corresponding feedback data.

[0125] For example, the number of preset machine learning models is 7, corresponding to each day of the calendar week. In this case, 7 historical data sets can be constructed to respectively correspond to the 7 preset machine learning models. At this time, the time data can be omitted from the historical data set.

[0126] Furthermore, as the number of times the household appliance is used increases, the amount of data in the historical data set increases accordingly, and the amounts of data included in the corresponding training set and validation set also continuously increase. For example, after each use of the washing machine, the newly collected data group is incorporated into the historical data set, and then the training set and validation set need to be reallocated to retrain and validate the preset machine learning model.

[0127] In a variant, the historical data set can be randomly divided into three subsets according to a certain ratio: a training set, a validation set, and a test set. After the updated preset machine learning model obtained by training based on the training set passes the validation through the validation set, it can be further finally debugged based on the test set. If the test passes, the updated preset machine learning model can be put into the daily use of the household appliance; otherwise, after reallocating to form new subsets, the training, validation, and test processes are repeated.

[0128] In a specific implementation, for the feedback data, the corresponding historical operation program, and time data received within a period of time, before constructing the training set and the validation set, the acquired data can be preprocessed first.

[0129] Specifically, the data preprocessing may include data cleaning to achieve a denoising effect. For example, abnormal data in the acquired data can be removed, such as the peak value in a continuous data segment. Another example is that error data, such as data with packet loss during transmission, can be removed.

[0130] Figure 4 It is a schematic diagram of a household appliance 1 according to an embodiment of the present invention.

[0131] Specifically, referring to Figure 4 , the household appliance 1 in this embodiment may include a main body 11.

[0132] Furthermore, the household appliance 1 may further include a control module 12, which is disposed on the main body 11. The control module 12 is used to execute the method for automatically setting the operation program of the household appliance described in the embodiment shown in Figures 1 to 3 .

[0133] Furthermore, the household appliance 1 may include an interactive panel to achieve human-computer interaction. The interactive panel may include a touch panel, a display screen, and the like.

[0134] The interactive panel may include a display and / or input unit 14. In some embodiments, the display and / or input unit 14 may be a user interface module (UIM) for displaying setting parameters or states of the household appliance and / or receiving control instructions input by a user.

[0135] Further, the interactive panel may also include a cover plate (not shown) arranged in front of the display and / or input unit 14 to at least play a protective role. The cover plate may be, for example, the front panel of the door 16 of the household appliance 1, or may be, for example, an independent glass plate covered in front of the UIM. In some embodiments, the display and / or input unit 14 may be close to the back side of the cover plate, so as to, for example, receive input signals input by the user via touching the cover plate. The display and / or input unit 14 may include a light-emitting member to allow light to pass through the corresponding area of the cover plate. In this embodiment, the front and rear direction refers to the depth direction of the household appliance 1, wherein the front or front side refers to the direction facing the user when the household appliance 1 is in use, and the rear or rear side refers to the direction away from the user when the household appliance 1 is in use.

[0136] Furthermore, the interactive panel may include a gesture sensing area for sensing and responding to gestures made by the user within a certain distance in front of the household appliance 1 to achieve gesture control. In some embodiments, the gesture sensing area may cover the entire interactive panel or may correspond to a portion of the interactive panel. For example, the preset trigger condition may be determined to be satisfied by detecting a user's gesture, and the gesture may be used to enable the automatic setting function of the running program of the household appliance 1.

[0137] Figure 4 Taking a washing machine as an example, the household appliance 1 is exemplified. For a washing machine with an automatic program setting function (also called an automatic program setting function), in order to more intelligently provide a suitable program according to the user's usage habits, in this embodiment, the control module 12 executes the above Figures 1 to 3 The method for automatically setting the operating program of a household appliance is shown. The method intelligently and automatically sets the operating program according to the different laundry needs of the user every day, thereby improving the user experience. Furthermore, the preset machine learning model for outputting the operating program can also be iteratively updated according to the time of each use, the operating program, and the satisfaction data fed back by the user.

[0138] Further, the household appliance 1 can be selected from: washing machines, washer-dryers, and dryers. The usage of such household appliances usually exhibits periodic patterns according to the user's work and living habits. Therefore, by learning the user's work and living patterns, the accuracy of predicting the operating program can be increased, further enhancing the user's satisfaction.

[0139] Thus, in view of the significant differences in the laundry styles of users at different times (e.g., weekdays and weekends), the method described in this implementation scheme is applied to household appliances 1 such as washing machines, enabling the washing machine to provide more intelligent washing and care functions for users. Further, in addition to washing machines, the method for automatically setting the operating program of the household appliances described in this implementation scheme can also be extended and applied to other washing / drying household appliances 1, making the entire laundry-to-drying process more intelligent and convenient.

[0140] In some embodiments, continuing to refer to Figure 4 , the household appliance 1 may further include: a communication module 13, disposed on the body 11, and the control module 12 receives feedback data through the communication module 13. Thus, the communication module 13 can implement communication in the form of, for example, wired or wireless. The household appliance 1 establishes a communication connection with the outside world (e.g., a server) through the communication module 13 to achieve remote update of the preset machine learning model.

[0141] Further, the communication module 13 can communicate with the control module 12 in a wired and / or wireless manner.

[0142] In some embodiments, the control module 12 can receive feedback data through the display and / or input unit 14. Further, prompt information can be sent through the display and / or input unit 14. Feedback data input by the user can also be received through the display and / or input unit 14. Thus, the user can transmit the satisfaction through network communication or directly input feedback information through the display and / or input unit 14, improving the convenience of feedback.

[0143] In some embodiments, the control module 12 can communicate with the user's terminal device 3 through the communication module 13. For example, the feedback data can be received from the terminal device 3 through the communication module 13.

[0144] It should be noted that Figure 4 only exemplarily shows the possible installation positions of the control module 12, communication module 13, display and / or input unit 14, power-on button 15, and door 16 on the body 11 of the household appliance 1. In actual applications, the mutual positional relationship of each module / component and the specific installation positions on the household appliance 1 can be adjusted according to needs.

[0145] Further, each module can be independent of each other, or integrated on the same chip or integrated into the same functional module. For example, the control module 12 and the communication module 13 can be integrated together.

[0146] Further, the modules can communicate with each other in a wired or wireless manner.

[0147] Figure 5 It is a schematic diagram of the principle of an automatic setting system for the operating program of a household appliance according to an embodiment of the present invention.

[0148] Specifically, referring to Figure 5 , the automatic setting system for the operating program of the household appliance described in this embodiment may include: a household appliance 1 (as shown in Figure 4 ), including a body 11 and a control module 12, and the control module 12 is used to execute the automatic setting method for the operating program of the household appliance described in the embodiment shown in Figures 1 to 3 ; a server 2, which communicates with the control module 12, and the server 2 is used to synchronize a preset machine learning model to the control module 12.

[0149] Further, the server 2 can be used to iterate and update the preset machine learning model based on the received historical operating program and feedback data of the household appliance 1, and synchronize the updated preset machine learning model to the control module 12.

[0150] In some embodiments, the server 2 can be integrated into the household appliance 1 to be dedicated to dynamically updating the preset machine learning model for the household appliance 1.

[0151] In some embodiments, the server 2 can be, for example, a background server, which is set at the manufacturer or designer of the household appliance 2. Or, the server 2 can be, for example, a cloud server. A single server 2 can communicate with multiple household appliances 1. For each household appliance 1, the server 2 receives the historical operating program and the corresponding feedback data of the household appliance 1 to dynamically update the preset machine learning model for the household appliance 1 specifically.

[0152] Thus, the server 2 can be used to store historical data, including the historical operating program and the corresponding feedback data, and retrain the model based on the historical data, so that the current operating program predicted by the updated model when the household appliance 1 is powered on next time is more in line with the user's usage habits at the current time. Further, when the server 2 is externally provided to the household appliance 1, it is beneficial to reduce the number of components in the household appliance 1 and reduce costs.

[0153] In some embodiments, the server 2 can be externally disposed relative to the household appliance 1 and communicate with the control module 12. For example, the server 2 can be set in the cloud and communicate with the control module 12 through a network. Thus, the server 2 can store more historical data and provide higher computing power for training and validating the preset machine learning model.

[0154] Continuing to refer Figure 5 , in some embodiments, the operating program automatic setting system of the household appliance described in this embodiment may further include: a communication module 13, disposed on the body 11, and the control module 12 communicates with the server 2 through the communication module 13.

[0155] Furthermore, the operating program automatic setting system of the household appliance described in this embodiment may further include: a terminal device 3, such as the user's mobile phone, tablet computer, other smart home appliances in the same local area network as the household appliance 1, etc. The terminal device 3 can communicate with the communication module 13 of the household appliance 1 or the server 2 through the built-in wireless network communication function.

[0156] In a specific implementation, the communication module 13 may include a WI-FI module, and the control module 12 remotely uploads the collected data (for example, time data, operating program data, and feedback data) to the server 2 through the WI-FI module.

[0157] Furthermore, the WI-FI module can also be used to upgrade the preset machine learning model via Over-The-Air (OTA) technology. For example, an OTA platform can be built at the server 2, and the updated preset machine learning model is updated via OTA through the WI-FI module.

[0158] Thus, the household appliance 1 establishes a communication connection with the outside world (for example, the server 2) through the communication module 13 to achieve remote update of the preset machine learning model.

[0159] Furthermore, the operating program automatic setting system of the household appliance described in this embodiment may further include: a display and / or input unit 14, disposed on the body 11 and communicating with the control module 12, and the display and / or input unit 14 is used to receive feedback data. Thus, the user can submit feedback data locally on the household appliance 1, and the household appliance 1 aggregates the corresponding historical operating programs together with the feedback data to the server 2 side, so that the server 2 iteratively updates the preset machine learning model based on these data.

[0160] In a typical application scenario, in combination with Figures 4 to 6 , taking the household appliance 1 as a washing machine as an example, the control module 12 of the washing machine can execute the above Figures 1 to 3For the method of the illustrated embodiment, the washing machine may include a storage module (not shown in the figure) for storing a preset machine learning model.

[0161] Specifically, the number of preset machine learning models is 7, corresponding to Monday to Sunday respectively. Different calendar days may correspond to different operating programs according to user habits. The operating program may include the specific washing program to be executed and related parameters.

[0162] In response to detecting that the washing machine is powered on, the control module 12 obtains the current time and inputs it into the corresponding preset machine learning model. Assuming the current time is Monday, the control module 12 may input Monday into the preset machine learning model corresponding to Monday to obtain the current operating program predicted by the preset machine learning model.

[0163] Then, the control module 12 sets each component of the washing machine to operate according to the predicted current operating program.

[0164] After the current operating program finishes running, the control module 12 issues a prompt message through the display and / or input unit 14 to prompt the user for feedback.

[0165] In response to receiving the feedback data, the control module 12 transmits the feedback data to the server 2 (e.g., cloud server) through the communication module 13. Further, the control module 12 also transmits the current time and the current operating program that triggered this feedback to the server 2 through the communication module 13.

[0166] In response to receiving the feedback data and the corresponding current time and current operating program, the server 2 may retrain the preset machine learning model corresponding to Monday based on the newly received data and the historical data related to Monday received historically.

[0167] Further, the communication module 13 of the washing machine communicates with the server 2 to remotely update the preset machine learning model.

[0168] In a variant, continuing to refer to Figure 6 , the specific process of iterative update of the preset machine learning model may further include:

[0169] First, the washing machine connects to the network through the WI-FI module and uploads the time data and operating program data collected by the washing machine to the cloud server 2.

[0170] Secondly, the user provides feedback on the performance of this laundry through the terminal device 3. The feedback data is stored in the server 2 in the cloud. Specifically, an opinion feedback form can be filled out through an application (Application, abbreviated as APP) installed on the terminal device 3 to generate feedback data. In response to receiving the feedback data and the historical operation program from the server 2 and the washing machine respectively, the server 2 can execute Figure 3 the steps shown to iteratively update the preset machine learning model based on the received data. Among them, the server 2 can match the feedback data and the historical operation program according to the number of times the data is received to determine the historical operation program corresponding to the feedback data. Further, each time the user provides feedback data, the server 2 can be triggered accordingly to perform an iterative update of the preset machine learning model.

[0171] Finally, the server 2 synchronizes the updated preset machine learning model to the washing machine through the OTA platform via the communication module 13 of the washing machine.

[0172] Although specific implementation manners have been described above, these implementation manners are not intended to limit the scope of the present disclosure, even in the case of describing a single implementation manner only with respect to specific features. The feature examples provided in the present disclosure are intended to be illustrative rather than restrictive, unless otherwise stated. In specific implementations, the technical features of one or more dependent claims can be combined with the technical features of the independent claim, and the technical features from the corresponding independent claims can be combined in any appropriate manner rather than only through the specific combinations listed in the claims.

[0173] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be subject to the scope defined by the claims.

Claims

1. A method for automatically setting an operating program of a household appliance, characterized in that including: Upon detecting that a preset trigger condition is satisfied, obtain the current time, where the preset trigger condition is used to trigger an automatic setting of the operating program; Input the current time into a preset machine learning model to obtain the current operating program, where the preset machine learning model is used to predict the current operating program based on the current time and the historical operating programs of the household appliance at related times in the past, and the related times are associated with the current time; Set the household appliance to operate according to the current operating program.

2. The method according to claim 1, wherein The related time being associated with the current time includes: the related time and the current time have a preset time interval, and the preset time interval includes calendar weeks.

3. The method according to claim 2, wherein Different times correspond to different preset machine learning models. The inputting the current time into a preset machine learning model to obtain the current operating program includes: Determine the corresponding preset machine learning model according to the current time; Input the current time into the determined preset machine learning model to obtain the current operating program.

4. The method according to claim 3, wherein The number of the preset machine learning models is 7, each corresponding to a different day of the calendar week.

5. The method according to claim 3, characterized in that, The preset machine learning models corresponding to different times are the same when the household appliance is powered on for the first time in its history, and change independently as the household appliance is used.

6. The method according to claim 1, characterized in that The preset machine learning model is iteratively updated as the number of times the household appliance is used increases.

7. The method according to claim 6, wherein The iterative update process of the preset machine learning model includes: Receive feedback data, where the feedback data is used to represent the satisfaction with the setting of the historical operating program; Based on the feedback data received within a period of time and the corresponding historical operating programs, construct a training set and a validation set; Train the preset machine learning model based on the training set to obtain an updated preset machine learning model; Validate the updated preset machine learning model based on the validation set.

8. The method according to claim 7, characterized in that The feedback data includes satisfaction scores for at least one parameter selected from: dryness, noise, entanglement degree, and the correctness of the automatic setting of the operating program.

9. The method according to claim 8, wherein, It further includes: Send a prompt message, where the prompt message includes a feedback form; Receive the feedback form and generate the feedback data based on the feedback form.

10. The method according to claim 9, wherein The prompt message and / or the feedback form are transmitted through the display and / or input unit of the household appliance, and / or, the prompt message and / or the feedback form are transmitted through a terminal device associated with the household appliance.

11. An household appliance, characterized in that, including: a body (11); a control module (12), disposed on the body (11), and the control module (12) is used to execute the method according to any one of claims 1 to 10 above.

12. The household appliance according to claim 11, characterized in that, The household appliance is selected from: washing machines, washer-dryers, and dryers.

13. The household appliance according to claim 11, characterized in that, It further includes: a communication module (13), disposed on the body (11), and the control module (12) receives feedback data through the communication module (13); and / or a display and / or input unit (14), disposed on the body (11), and the control module (12) receives feedback data through the display and / or input unit (14).

14. An operating program automatic setting system for a household appliance, characterized in that, Comprising: A household appliance (1), comprising a main body (11) and a control module (12), the control module (12) being configured to execute the method according to any one of claims 1 to 10 above; A server (2), communicating with the control module (12), the server (2) being configured to synchronize the preset machine learning model to the control module (12).

15. The system according to claim 14, wherein The server (2) is configured to iteratively update the preset machine learning model based on the historical operation program and feedback data of the received household appliance (1), and synchronize the updated preset machine learning model to the control module (12).

Images